<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Developer-Tools on Quentin Rousseau</title><link>http://blog.quent.in/tags/developer-tools/</link><description>Recent content in Developer-Tools on Quentin Rousseau</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Tue, 19 May 2026 08:00:00 +0000</lastBuildDate><atom:link href="http://blog.quent.in/tags/developer-tools/index.xml" rel="self" type="application/rss+xml"/><item><title>Your Small PR Rule Won't Survive AI</title><link>http://blog.quent.in/blog/2026/05/19/your-small-pr-rule-wont-survive-ai/</link><pubDate>Tue, 19 May 2026 08:00:00 +0000</pubDate><guid>http://blog.quent.in/blog/2026/05/19/your-small-pr-rule-wont-survive-ai/</guid><description>&lt;p&gt;For two years, we enforced a strict small-PR culture at &lt;a href="https://rootly.com"target="_blank"
 class="inline-flex items-center gap-1"
&gt;Rootly&lt;svg class="h-3 w-3 flex-shrink-0" id="external-link" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"&gt;&lt;path fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M15 3h6v6m-11 5L21 3m-3 10v6a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h6"/&gt;&lt;/svg&gt;
&lt;/a&gt;. Stacked PRs, atomic changes, never more than a couple hundred lines. The logic was sound: smaller diffs are easier to review, easier to revert, easier to reason about.&lt;/p&gt;
&lt;p&gt;Then AI started writing most of our code.&lt;/p&gt;</description></item><item><title>One More Prompt: The Dopamine Trap of Agentic Coding</title><link>http://blog.quent.in/blog/2026/03/09/one-more-prompt-the-dopamine-trap-of-agentic-coding/</link><pubDate>Mon, 09 Mar 2026 14:00:00 +0000</pubDate><guid>http://blog.quent.in/blog/2026/03/09/one-more-prompt-the-dopamine-trap-of-agentic-coding/</guid><description>&lt;p&gt;It&amp;rsquo;s 2:47 AM on a Tuesday. I&amp;rsquo;m not debugging a production outage. There&amp;rsquo;s no deadline. I&amp;rsquo;m just watching Claude Code refactor a module — and I can&amp;rsquo;t stop. One more prompt. One more agent run. One more hit.&lt;/p&gt;
&lt;p&gt;If this sounds familiar, you&amp;rsquo;re not alone. Something strange is happening to developers in 2026: &lt;strong&gt;AI coding tools may be triggering the same dopamine loops as slot machines&lt;/strong&gt;, and the tech industry is sleepwalking into it.&lt;/p&gt;</description></item><item><title>From Writing Code to Orchestrating Agents</title><link>http://blog.quent.in/blog/2026/03/09/from-vs-code-to-agent-orchestrators-how-multi-agent-workflows-changed-everything/</link><pubDate>Mon, 09 Mar 2026 08:47:29 +0000</pubDate><guid>http://blog.quent.in/blog/2026/03/09/from-vs-code-to-agent-orchestrators-how-multi-agent-workflows-changed-everything/</guid><description>&lt;p&gt;My dev workflow has changed more in early 2026 than in the previous five years. I went from writing code in VS Code, to pair-programming with AI in Cursor, to running CLI agents like Claude Code and Codex — and now I&amp;rsquo;m orchestrating multiple agents working in parallel on different tasks at the same time.&lt;/p&gt;
&lt;p&gt;Each step felt like a leap. But the jump to multi-agent orchestration is the one that fundamentally changed how I think about shipping software.&lt;/p&gt;</description></item><item><title>Your AI Agent Doesn’t Need an MCP Server — It Needs a Good CLI</title><link>http://blog.quent.in/blog/2026/02/24/your-ai-agent-doesnt-need-an-mcp-server-it-needs-a-good-cli/</link><pubDate>Tue, 24 Feb 2026 23:27:05 +0000</pubDate><guid>http://blog.quent.in/blog/2026/02/24/your-ai-agent-doesnt-need-an-mcp-server-it-needs-a-good-cli/</guid><description>&lt;p&gt;Every week, a new MCP server shows up on GitHub that wraps a CLI tool that already existed. MCP server for GitHub? &lt;code&gt;gh&lt;/code&gt; has been doing that since 2020. MCP server for AWS? The &lt;code&gt;aws&lt;/code&gt; CLI covers 200+ services with &lt;code&gt;--output json&lt;/code&gt;. MCP server for Datadog? They shipped a CLI instead — and it works better.&lt;/p&gt;
&lt;p&gt;I get the excitement around &lt;a href="https://modelcontextprotocol.io/"target="_blank"
 class="inline-flex items-center gap-1"
&gt;MCP&lt;svg class="h-3 w-3 flex-shrink-0" id="external-link" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"&gt;&lt;path fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M15 3h6v6m-11 5L21 3m-3 10v6a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h6"/&gt;&lt;/svg&gt;
&lt;/a&gt; (Model Context Protocol). It&amp;rsquo;s a clean standard for connecting AI to external systems. Anthropic designed it well. But somewhere along the way, the community started treating MCP as the &lt;em&gt;default&lt;/em&gt; way to give AI agents access to tools — even when a perfectly good CLI already exists.&lt;/p&gt;</description></item><item><title>From "Maintainers Wanted" to Three Major Releases in a Weekend</title><link>http://blog.quent.in/blog/2026/02/24/from-maintainers-wanted-to-three-major-releases-in-a-weekend/</link><pubDate>Tue, 24 Feb 2026 22:42:22 +0000</pubDate><guid>http://blog.quent.in/blog/2026/02/24/from-maintainers-wanted-to-three-major-releases-in-a-weekend/</guid><description>&lt;p&gt;I have 50+ public repositories on GitHub. Some of them haven&amp;rsquo;t been touched in years. A few still get stars and forks weekly. Most sit somewhere in between — not abandoned, not active, just&amp;hellip; there.&lt;/p&gt;
&lt;p&gt;Until recently, that &amp;ldquo;just there&amp;rdquo; state came with guilt. Every Dependabot PR I didn&amp;rsquo;t merge, every issue I didn&amp;rsquo;t respond to, every README that still referenced a deprecated API — it all felt like debt accumulating. Maintaining open-source repositories was a second job I never signed up for.&lt;/p&gt;</description></item></channel></rss>